In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
Make sure that you've downloaded the required human and dog datasets:
Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dogImages.
Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.
Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.
In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.
import numpy as np
from glob import glob
# load filenames for human and dog images
human_files = np.array(glob("lfw/*/*"))
dog_files = np.array(glob("dogImages/*/*/*"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 13233 total human images. There are 8351 total dog images.
In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.
OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
import cv2
import matplotlib.pyplot as plt
%matplotlib inline
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Number of faces detected: 1
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
Question 1: Use the code cell below to test the performance of the face_detector function.
human_files have a detected human face? dog_files have a detected human face? Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer:
from tqdm import tqdm
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
## TODO: Test the performance of the face_detector algorithm
## on the images in human_files_short and dog_files_short.
face_in_human = 0
for i in tqdm(range(len(human_files_short))):
if face_detector(human_files_short[i]):
face_in_human += 1
face_in_dog = 0
for i in tqdm(range(len(dog_files_short))):
if face_detector(dog_files_short[i]):
face_in_dog += 1
print(f"Found {face_in_human} faces [{(face_in_human/len(human_files_short))*100}%] in human_files_short")
print(f"Found {face_in_dog} faces [{(face_in_dog/len(dog_files_short))*100}%] in dog_files_short")
100%|████████████████████████████████████████████████████████████████████████████████| 100/100 [00:04<00:00, 20.15it/s] 100%|████████████████████████████████████████████████████████████████████████████████| 100/100 [00:22<00:00, 4.41it/s]
Found 96 faces [96.0%] in human_files_short Found 18 faces [18.0%] in dog_files_short
We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Test performance of another face detection algorithm.
### Feel free to use as many code cells as needed.
In this section, we use a pre-trained model to detect dogs in images.
The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.
import torch
import torchvision.models as models
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
print("Cuda is available")
VGG16 = VGG16.cuda()
Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.
In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.
Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.
from PIL import Image
import torchvision.transforms as transforms
# Set PIL to be tolerant of image files that are truncated.
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
def VGG16_predict(img_path):
'''
Use pre-trained VGG-16 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to VGG-16 model's prediction
'''
## TODO: Complete the function.
## Load and pre-process an image from the given img_path
## Return the *index* of the predicted class for that image
image = Image.open(img_path)
min_img_size = 224
# The min size, as noted in the PyTorch pretrained models doc, is 224 px.
transform = transforms.Compose([transforms.Resize(min_img_size), transforms.CenterCrop(min_img_size),
transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])])
image = transform(image)
image = image.unsqueeze(0)
if use_cuda:
image = image.to('cuda')
## Return the *index* of the predicted class for that image
output = VGG16(image)
if use_cuda:
output = output.to('cpu')
return output.data.numpy().argmax()# predicted class index
VGG16_predict('dogImages/train/006.American_eskimo_dog/American_eskimo_dog_00409.jpg')
258
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).
Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
## TODO: Complete the function.
prediction = VGG16_predict(img_path)
if(prediction > 150 and prediction < 269):
return True
return False # true/false
Question 2: Use the code cell below to test the performance of your dog_detector function.
human_files_short have a detected dog? dog_files_short have a detected dog?Answer:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
dogs_in_human = 0
dogs_in_dogs = 0
for i in tqdm (range(len(human_files_short))):
if dog_detector(human_files_short[i]):
dogs_in_human += 1
for i in tqdm (range(len(dog_files_short))):
if dog_detector(dog_files_short[i]):
dogs_in_dogs += 1
print(f"Found {dogs_in_human} dogs [{(dogs_in_human/len(human_files_short))*100}%] in human_files")
print(f"Found {dogs_in_dogs} dogs [{(dogs_in_dogs/len(dog_files_short))*100}%] in dog_files")
100%|████████████████████████████████████████████████████████████████████████████████| 100/100 [01:06<00:00, 1.51it/s] 100%|████████████████████████████████████████████████████████████████████████████████| 100/100 [01:06<00:00, 1.49it/s]
Found 0 dogs [0.0%] in human_files Found 94 dogs [94.0%] in dog_files
We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
![]() |
![]() |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
![]() |
![]() |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
![]() |
![]() |
![]() |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!
import os
from torchvision import datasets
### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes
batch_size = 20
min_img_size = 224
transform_train = transforms.Compose([transforms.RandomResizedCrop(min_img_size), transforms.RandomRotation(10),
transforms.RandomHorizontalFlip(), transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])
transform_valid = transforms.Compose([transforms.Resize(min_img_size), transforms.CenterCrop(min_img_size),
transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])])
train_path = datasets.ImageFolder('dogImages/train',transform=transform_train)
test_path = datasets.ImageFolder('dogImages/test',transform=transform_valid)
valid_path = datasets.ImageFolder('dogImages/valid',transform=transform_valid)
loaders_scratch = {'train': torch.utils.data.DataLoader(train_path, batch_size=batch_size, shuffle=True),
'test': torch.utils.data.DataLoader(test_path, batch_size=batch_size, shuffle=True),
'valid': torch.utils.data.DataLoader(valid_path, batch_size=batch_size, shuffle=True)}
Question 3: Describe your chosen procedure for preprocessing the data.
Answer:
Create a CNN to classify dog breed. Use the template in the code cell below.
import torch.nn as nn
import torch.nn.functional as F
# define the CNN architecture
class Net(nn.Module):
### TODO: choose an architecture, and complete the class
def __init__(self):
super(Net, self).__init__()
## Define layers of a CNN
# 4 Convolutional layers
self.conv_1 = nn.Conv2d(3,16,3,padding=1)
self.conv_2 = nn.Conv2d(16,32,3,padding=1)
self.conv_3 = nn.Conv2d(32,64,3,padding=1)
self.conv_4 = nn.Conv2d(64,128,3,padding=1)
self.conv1_bn = nn.BatchNorm2d(16)
self.conv2_bn = nn.BatchNorm2d(32)
self.conv3_bn = nn.BatchNorm2d(64)
self.conv4_bn = nn.BatchNorm2d(128)
self.pool = nn.MaxPool2d(2, 2)
# 3 Linear layers
# linear layer (128 * 14 * 14 -> 12544)
self.fc_1 = nn.Linear(128 * 14 * 14, 1024)
# linear layer (12544 -> 6272)
self.fc_2 = nn.Linear(1024, 512)
# linear layer (12544 -> 6272)
self.fc_3 = nn.Linear(512, 133)
# dropout layer (p=0.25)
self.dropout = nn.Dropout(0.5)
def forward(self, x):
## Define forward behavior
x = self.pool(F.relu(self.conv1_bn(self.conv_1(x))))
x = self.pool(F.relu(self.conv2_bn(self.conv_2(x))))
x = self.pool(F.relu(self.conv3_bn(self.conv_3(x))))
x = self.pool(F.relu(self.conv4_bn(self.conv_4(x))))
# flatten image input
x = x.view(-1, np.product(x.shape[1:]))
# add dropout layer
x = self.dropout(x)
# 1st hidden layer, with relu activation function
x = F.relu(self.fc_1(x))
# add dropout layer
x = self.dropout(x)
# 2nd hidden layer, with relu activation function
x = F.relu(self.fc_2(x))
# add dropout layer
x = self.dropout(x)
# output layer with softmax
x = self.fc_3(x)
return x
#-#-# You do NOT have to modify the code below this line. #-#-#
# instantiate the CNN
model_scratch = Net()
# move tensors to GPU if CUDA is available
if use_cuda:
model_scratch.cuda()
Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.
Answer:
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.
import torch.optim as optim
### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()
### TODO: select optimizer
optimizer_scratch = optim.Adam(model_scratch.parameters(), lr=0.001)
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.
from PIL import ImageFile
import time
ImageFile.LOAD_TRUNCATED_IMAGES = True
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
"""returns trained model"""
# initialize tracker for minimum validation loss
valid_loss_min = np.Inf
for epoch in range(1, n_epochs+1):
# initialize variables to monitor training and validation loss
train_loss = 0.0
valid_loss = 0.0
start_time = time.time()
###################
# train the model #
###################
model.train()
for batch_idx, (data, target) in enumerate(loaders['train']):
#print(f'Path: {path}\n')
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
## find the loss and update the model parameters accordingly
## record the average training loss, using something like
## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
optimizer.zero_grad()
output = model(data)
loss = criterion(output, target)
loss.backward()
optimizer.step()
train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
######################
# validate the model #
######################
model.eval()
for batch_idx, (data, target) in enumerate(loaders['valid']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
## update the average validation loss
output = model(data)
loss = criterion(output,target)
valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))
# print training/validation statistics
end_time = time.time()
epoch_time = end_time - start_time
print('Epoch: {} \tSec. spent in epoch {:.2f}\tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
epoch,
epoch_time,
train_loss,
valid_loss
))
## TODO: save the model if validation loss has decreased
if valid_loss <= valid_loss_min:
print('Validation loss decreased ({:.46} --> {:.6f}). Saving model'.format(
valid_loss_min,
valid_loss))
torch.save(model.state_dict(), save_path)
valid_loss_min = valid_loss
# return trained model
return model
# train the model
model_scratch = train(40, loaders_scratch, model_scratch, optimizer_scratch,
criterion_scratch, use_cuda, 'model_scratch.pt')
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
Epoch: 1 Sec. spent in epoch 1007.81 Training Loss: 5.045363 Validation Loss: 4.867544 Validation loss decreased (inf --> 4.867544). Saving model Epoch: 2 Sec. spent in epoch 1733.52 Training Loss: 4.881907 Validation Loss: 4.858373 Validation loss decreased (4.867543697357177734375 --> 4.858373). Saving model Epoch: 3 Sec. spent in epoch 1049.54 Training Loss: 4.858354 Validation Loss: 4.813962 Validation loss decreased (4.85837268829345703125 --> 4.813962). Saving model Epoch: 4 Sec. spent in epoch 1097.96 Training Loss: 4.832860 Validation Loss: 4.776076 Validation loss decreased (4.81396198272705078125 --> 4.776076). Saving model Epoch: 5 Sec. spent in epoch 1118.09 Training Loss: 4.818565 Validation Loss: 4.782058 Epoch: 6 Sec. spent in epoch 1103.88 Training Loss: 4.807873 Validation Loss: 4.743745 Validation loss decreased (4.776075839996337890625 --> 4.743745). Saving model Epoch: 7 Sec. spent in epoch 1111.70 Training Loss: 4.794950 Validation Loss: 4.747508 Epoch: 8 Sec. spent in epoch 1114.42 Training Loss: 4.786737 Validation Loss: 4.700951 Validation loss decreased (4.74374485015869140625 --> 4.700951). Saving model Epoch: 9 Sec. spent in epoch 1129.31 Training Loss: 4.777702 Validation Loss: 4.712528 Epoch: 10 Sec. spent in epoch 1111.26 Training Loss: 4.778144 Validation Loss: 4.685511 Validation loss decreased (4.700951099395751953125 --> 4.685511). Saving model Epoch: 11 Sec. spent in epoch 1130.35 Training Loss: 4.759265 Validation Loss: 4.700334 Epoch: 12 Sec. spent in epoch 1114.96 Training Loss: 4.740970 Validation Loss: 4.680397 Validation loss decreased (4.6855106353759765625 --> 4.680397). Saving model Epoch: 13 Sec. spent in epoch 1170.92 Training Loss: 4.724675 Validation Loss: 4.641817 Validation loss decreased (4.680396556854248046875 --> 4.641817). Saving model Epoch: 14 Sec. spent in epoch 1126.19 Training Loss: 4.703886 Validation Loss: 4.630018 Validation loss decreased (4.6418170928955078125 --> 4.630018). Saving model Epoch: 15 Sec. spent in epoch 1132.28 Training Loss: 4.692550 Validation Loss: 4.617931 Validation loss decreased (4.6300182342529296875 --> 4.617931). Saving model Epoch: 16 Sec. spent in epoch 1064.57 Training Loss: 4.681177 Validation Loss: 4.604334 Validation loss decreased (4.617930889129638671875 --> 4.604334). Saving model Epoch: 17 Sec. spent in epoch 1049.38 Training Loss: 4.656940 Validation Loss: 4.589087 Validation loss decreased (4.6043338775634765625 --> 4.589087). Saving model Epoch: 18 Sec. spent in epoch 1046.38 Training Loss: 4.619784 Validation Loss: 4.467548 Validation loss decreased (4.58908748626708984375 --> 4.467548). Saving model Epoch: 19 Sec. spent in epoch 1050.42 Training Loss: 4.580341 Validation Loss: 4.402242 Validation loss decreased (4.467548370361328125 --> 4.402242). Saving model Epoch: 20 Sec. spent in epoch 1031.60 Training Loss: 4.526961 Validation Loss: 4.314043 Validation loss decreased (4.402242183685302734375 --> 4.314043). Saving model Epoch: 21 Sec. spent in epoch 1045.16 Training Loss: 4.469679 Validation Loss: 4.289766 Validation loss decreased (4.314042568206787109375 --> 4.289766). Saving model Epoch: 22 Sec. spent in epoch 1036.87 Training Loss: 4.451042 Validation Loss: 4.291351 Epoch: 23 Sec. spent in epoch 1036.85 Training Loss: 4.434579 Validation Loss: 4.237468 Validation loss decreased (4.289765834808349609375 --> 4.237468). Saving model Epoch: 24 Sec. spent in epoch 1033.99 Training Loss: 4.410046 Validation Loss: 4.256171 Epoch: 25 Sec. spent in epoch 1030.23 Training Loss: 4.373800 Validation Loss: 4.203207 Validation loss decreased (4.237468242645263671875 --> 4.203207). Saving model Epoch: 26 Sec. spent in epoch 1043.36 Training Loss: 4.352248 Validation Loss: 4.191356 Validation loss decreased (4.203207492828369140625 --> 4.191356). Saving model Epoch: 27 Sec. spent in epoch 1038.84 Training Loss: 4.348096 Validation Loss: 4.184841 Validation loss decreased (4.191356182098388671875 --> 4.184841). Saving model Epoch: 28 Sec. spent in epoch 1003.33 Training Loss: 4.334260 Validation Loss: 4.185010 Epoch: 29 Sec. spent in epoch 968.89 Training Loss: 4.306895 Validation Loss: 4.130088 Validation loss decreased (4.184841156005859375 --> 4.130088). Saving model Epoch: 30 Sec. spent in epoch 974.50 Training Loss: 4.314743 Validation Loss: 4.122968 Validation loss decreased (4.13008785247802734375 --> 4.122968). Saving model Epoch: 31 Sec. spent in epoch 959.28 Training Loss: 4.281910 Validation Loss: 4.059975 Validation loss decreased (4.122968196868896484375 --> 4.059975). Saving model Epoch: 32 Sec. spent in epoch 973.62 Training Loss: 4.266283 Validation Loss: 4.048931 Validation loss decreased (4.059975147247314453125 --> 4.048931). Saving model Epoch: 33 Sec. spent in epoch 961.35 Training Loss: 4.256562 Validation Loss: 4.071686 Epoch: 34 Sec. spent in epoch 956.29 Training Loss: 4.246737 Validation Loss: 4.023552 Validation loss decreased (4.048931121826171875 --> 4.023552). Saving model Epoch: 35 Sec. spent in epoch 951.39 Training Loss: 4.216971 Validation Loss: 4.008220 Validation loss decreased (4.02355194091796875 --> 4.008220). Saving model Epoch: 36 Sec. spent in epoch 956.14 Training Loss: 4.185484 Validation Loss: 3.989025 Validation loss decreased (4.008220195770263671875 --> 3.989025). Saving model Epoch: 37 Sec. spent in epoch 949.74 Training Loss: 4.196458 Validation Loss: 3.957479 Validation loss decreased (3.989025115966796875 --> 3.957479). Saving model Epoch: 38 Sec. spent in epoch 940.44 Training Loss: 4.180910 Validation Loss: 3.973938 Epoch: 39 Sec. spent in epoch 936.46 Training Loss: 4.165804 Validation Loss: 3.961264 Epoch: 40 Sec. spent in epoch 956.91 Training Loss: 4.169381 Validation Loss: 3.954540 Validation loss decreased (3.95747852325439453125 --> 3.954540). Saving model
<All keys matched successfully>
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.
def test(loaders, model, criterion, use_cuda):
# monitor test loss and accuracy
test_loss = 0.
correct = 0.
total = 0.
model.eval()
for batch_idx, (data, target) in enumerate(loaders['test']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the loss
loss = criterion(output, target)
# update average test loss
test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
# convert output probabilities to predicted class
pred = output.data.max(1, keepdim=True)[1]
# compare predictions to true label
correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
total += data.size(0)
print('Test Loss: {:.6f}\n'.format(test_loss))
print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
100. * correct / total, correct, total))
# call test function
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Test Loss: 3.918597 Test Accuracy: 8% (72/836)
You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).
If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.
## TODO: Specify data loaders
transfer_loaders = loaders_scratch.copy()
Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.
import torchvision.models as models
import torch.nn as nn
## TODO: Specify model architecture
model_transfer = models.resnet50(pretrained=True)
# Freeze parameters so we don't backprop through them
for param in model_transfer.parameters():
param.requires_grad = False
# Replace the last fully connected layer with a Linnear layer with 133 out features
model_transfer.fc = nn.Linear(2048, 133)
if use_cuda:
model_transfer = model_transfer.cuda()
print(model_transfer)
ResNet(
(conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
(layer1): Sequential(
(0): Bottleneck(
(conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(downsample): Sequential(
(0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(2): Bottleneck(
(conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
(layer2): Sequential(
(0): Bottleneck(
(conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(downsample): Sequential(
(0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(2): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(3): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
(layer3): Sequential(
(0): Bottleneck(
(conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(downsample): Sequential(
(0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(2): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(3): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(4): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(5): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
(layer4): Sequential(
(0): Bottleneck(
(conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(downsample): Sequential(
(0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(2): Bottleneck(
(conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
(avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
(fc): Linear(in_features=2048, out_features=133, bias=True)
)
Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.
Answer:
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.Adam(model_transfer.fc.parameters(), lr=0.001)
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.
# train the model
model_transfer = train(10, transfer_loaders, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
# train(n_epochs, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
# load the model that got the best validation accuracy (uncomment the line below)
#model_transfer.load_state_dict(torch.load('model_transfer.pt'))
Epoch: 1 Sec. spent in epoch 2139.72 Training Loss: 2.699786 Validation Loss: 0.951117 Validation loss decreased (inf --> 0.951117). Saving model Epoch: 2 Sec. spent in epoch 2060.82 Training Loss: 1.423272 Validation Loss: 0.631965 Validation loss decreased (0.951116740703582763671875 --> 0.631965). Saving model Epoch: 3 Sec. spent in epoch 2031.03 Training Loss: 1.261887 Validation Loss: 0.695990 Epoch: 4 Sec. spent in epoch 2085.20 Training Loss: 1.218193 Validation Loss: 0.667064 Epoch: 5 Sec. spent in epoch 2079.26 Training Loss: 1.121667 Validation Loss: 0.571727 Validation loss decreased (0.631964504718780517578125 --> 0.571727). Saving model Epoch: 6 Sec. spent in epoch 2199.12 Training Loss: 1.113447 Validation Loss: 0.547890 Validation loss decreased (0.57172691822052001953125 --> 0.547890). Saving model Epoch: 7 Sec. spent in epoch 8956.73 Training Loss: 1.064585 Validation Loss: 0.620886 Epoch: 8 Sec. spent in epoch 2927.90 Training Loss: 1.064403 Validation Loss: 0.576969 Epoch: 9 Sec. spent in epoch 2418.39 Training Loss: 1.041807 Validation Loss: 0.530868 Validation loss decreased (0.547890186309814453125 --> 0.530868). Saving model Epoch: 10 Sec. spent in epoch 2132.96 Training Loss: 1.040170 Validation Loss: 0.520499 Validation loss decreased (0.530868113040924072265625 --> 0.520499). Saving model
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.
test(transfer_loaders, model_transfer, criterion_transfer, use_cuda)
Test Loss: 0.553724 Test Accuracy: 83% (701/836)
Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.
# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in train_path.classes]
def predict_breed_transfer(img_path):
# load the image and return the predicted breed
image = Image.open(img_path)
min_img_size = 224
# The min size, as noted in the PyTorch pretrained models doc, is 224 px.
transform = transforms.Compose([transforms.Resize(min_img_size), transforms.CenterCrop(min_img_size), transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])
image = transform(image)
image = image.unsqueeze(0)
if use_cuda:
image = image.to('cuda')
## Return the *index* of the predicted class for that image
output = model_transfer(image)
if use_cuda:
output = output.to('cpu')
return class_names[output.data.numpy().argmax()]# predicted class index
Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,
You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and dog_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.
Some sample output for our algorithm is provided below, but feel free to design your own user experience!

### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.
def run_app(img_path):
## handle cases for a human face, dog, and neither
img = Image.open(img_path)
plt.imshow(img)
plt.show()
if dog_detector(img_path) is True:
prediction = predict_breed_transfer(img_path)
print("Dog Detected!\nIt looks like a {0}".format(prediction))
elif face_detector(img_path) > 0:
prediction = predict_breed_transfer(img_path)
print("Hello, human!\nIf you were a dog... You look like a {0}".format(prediction))
else:
print("Urm... Are you an alien?")
# Load custom test images
human_files = np.array(glob("./lfw/Testing/*"))
dog_files = np.array(glob("./dogImages/test/Testing/*"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 15 total human images. There are 15 total dog images.
In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?
Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.
Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer:
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.
## suggested code, below
for file in np.hstack((human_files[:15], dog_files[:15])):
run_app(file)
Hello, human! If you were a dog... You look like a Dogue de bordeaux
Hello, human! If you were a dog... You look like a Bulldog
Hello, human! If you were a dog... You look like a Dogue de bordeaux
Hello, human! If you were a dog... You look like a Chinese crested
Hello, human! If you were a dog... You look like a Dogue de bordeaux
Hello, human! If you were a dog... You look like a Belgian malinois
Hello, human! If you were a dog... You look like a Dogue de bordeaux
Hello, human! If you were a dog... You look like a Dogue de bordeaux
Hello, human! If you were a dog... You look like a American water spaniel
Hello, human! If you were a dog... You look like a Great dane
Hello, human! If you were a dog... You look like a Doberman pinscher
Hello, human! If you were a dog... You look like a Doberman pinscher
Hello, human! If you were a dog... You look like a Dogue de bordeaux
Hello, human! If you were a dog... You look like a Dogue de bordeaux
Hello, human! If you were a dog... You look like a Dogue de bordeaux
Dog Detected! It looks like a Beagle
Dog Detected! It looks like a Belgian tervuren
Dog Detected! It looks like a Bluetick coonhound
Dog Detected! It looks like a Bouvier des flandres
Dog Detected! It looks like a Bulldog
Dog Detected! It looks like a Lakeland terrier
Dog Detected! It looks like a Miniature schnauzer
Dog Detected! It looks like a Miniature schnauzer
Dog Detected! It looks like a Old english sheepdog
Dog Detected! It looks like a Papillon
Dog Detected! It looks like a Parson russell terrier
Dog Detected! It looks like a Pembroke welsh corgi
Dog Detected! It looks like a Pembroke welsh corgi
Dog Detected! It looks like a Cardigan welsh corgi
Dog Detected! It looks like a Pomeranian